Brian's Brain
Created by Brian Silverman [EE], Brian's brain is a CA inspired by neural architecture that generates interesting global patterns.
Simplistically, we can view a brain as a CA. In this case cells represent neurons. Two neurons are neighbors if they are connected. A neuron is either firing, ready to fire, or recharging. When a cell is in its ready state, it fires when the number of neighboring cells firing (i.e. the ambience) is equal to some threshold value. Taking this threshold to be two creates interesting persistent patterns similar to Gosper gliders:
The unlabeled arrows indicate an unconditional transfer that happens in the next cycle. For example, if a cell is in the firing state, then on the next cycle it automatically transitions into the recharging state. In the next cycle it transitions into the ready state.
From the ready state it transitions into the firing state if the ambience is two. Ambience is defined as the number of neighbors in the on state.
The snapshots below show a simple repeating pattern in Brian's Brain:
